20 November 2011

Nordea’s ROE Target: Realistic, Optimistic or Irresponsible?

Nordea, one of the Europe's top ten largest universal banks in terms of total market capitalisation, states in its Q3 2011 report that it has replaced its previous financial targets with one: an ROE of 15% in a normalised macroeconomic environment. Needless to say that in light of the bank’s ROE of 6.5% in Q3 this target has been greeted with scepticism (is it realistic?) as well as with the opposition by the unions like Finansförbundet, the Financial Sector Union of Sweden (because of the plans to cut 2,000+ jobs).

I do not share the opinions of those who think that Swedish government should step in and use its key holding in Nordea to get the bank to lower its 15% return on equity target for preserving jobs. Yes, it may look unfair to some, but maintaining operational inefficiencies would sound like nonsense. However, 15% might be considered as irresponsibly high target because it is higher than the the average return on equity for non-financial companies in Sweden (see Figure 1 below). Of course, if one compares finance with gambling (remember casino banking?), then 15% is very low indeed...


15% ROE for a bank is not a responsible target from the financial stability point of view either. This conclusion derives from the very logic of money creation and interests. Suppose that all Swedish banks target a ROE of 15% (which they are doing, even if not saying so). As of 30 September 2011, the monetary base for Sweden was SEK 116.7bn. The total banks’ SEK equity was SEK 259.2bn; 15% of this amount is approximately SEK 38.9bn. This SEK 38.9bn or 33.3% of the monetary base is the money that has to be collected from the customers, but has never been created... (Base data: Sveriges Riksbank)

When it comes to the realism of the targeted ROE, then the first question apparently is: what’s the “normalised macroeconomic environment”? To this question, Nordea’s CFO Fredrik Rystedt provided an answer in his presentation in the Nordea’s Capital Markets’ Day 2011:
* Short-term interest rates 2-4%; long-term interest rates 3-5%;
* 4-6% asset appreciation in European stock markets;
* 4-6% business volume growth.
This looks pretty ok, except that there is no reasonable idea whatsoever about when we are going to see such normalised macroeconomic environment.

To summarise the Nordea’s guidance for achieving this 15% ROE target:
1. Reduction of staff by 2,000+ (6% of full time employees) will enable Nordea to keep costs unchanged for a prolonged period of time;
2. Improving efficiency in IT and operations;
3. (1) and (2) plus the assumption of the normalised economic environment would result in cost-income ratio < 48% (the starting point is the cost-income ratio of 68% in Q3 2011 / 54% in Q2 2011), and in a 2% increase in ROE;
4. Improving capital efficiency, i.e. reducing risk-weighted assets (RWA) and hence capital requirement;
5. (4) is based on the assumptions that the new regulatory requirements would add EUR 13-18bn to the current (Q3 2011) RWA of EUR 180bn, whereas improvements in the credit portfolio management, and models and process refinements would reduce it by EUR 21-27bn; this would result in RWA reduction to the level of EUR 170bn as of Q3 2011, i.e. give room for increasing business volumes.
6. The regulatory Common Equity Tier 1 (CET1) capital ratio is assumed to be 11%.
7. A fully Basel III compliant liquidity buffer is assumed to add approx EUR 40m of cost.
(Source: Fredrik Rystedt’ presentation in the Nordea’s Capital Markets’ Day 2011)

There appear to be some omissions (partly because of the new information that wasn’t known for sure at the point of presentation) and several question marks in the above guidance. Consider the following.

First, Nordea is going to need more equity than it has indicated so far:
* Having been identified as G-SIFI (Global Systematically Important Financial Institution) by BIS means an additional capital requirement 1% to 2.5% of risk-weighted assets.
* The assumed Q3 2011 RWA of EUR 180bn is the RWA without taking into account the transition rules (when taking into account the transition rules, the RWA was EUR 220bn). It’s not sure at all when the transition rules will be removed in Sweden. Furthermore, if it should happen one day, it’s still highly unlikely that the current very low risk weights can be used for the calculation of the capital need for Swedish banks. (As a reference, consider what the Riksbank’s Governor Stefan Ingves is saying about the implementation of Basel III in Sweden; the same message appears in each such speech.)

Secondly, having a considerable trading book (more precisely EUR 253.8bn or 37.8% of total assets as of 30 September 2011) is very likely to turn out (much) more expensive than assumed so far. For one thing, large trading book has negative impact to the perceived risks and the expected earnings’ volatility (which the analysts just saw in the Q3 results). Among others, this is also reflected in the S&P’s updated Banks Rating Methodology (rating implications). “Normalisation” in this portfolio is going to take rather long as troubling European banks are trying to get rid from at least part of their mark-to-market/mark-to-model assets.

Thirdly, funding costs are going to be higher than possibly assumed. Despite of all the explanations about the sound history of covered bonds and the depth of USD markets, high loan-to-deposit ratios (224% in Denmark and 209% in Sweden as of 30 September 2011) and the gap between the USD assets and liabilities (USD liabilities exceeded USD assets by EUR 37.8bn as of 30 September 2011) will have an impact, sooner or later. To illustrate the matter, consider the developments of the Danish Mortgage System. Covered bonds there today are no more these simple safe instruments that they once used to be... It’s only a matter of time when funding markets will price it in.

Fourthly, nothing specific has been pointed out about the “normalisation” of credit losses. I tend to think that this process will take rather long. Just to mention one thing: the remarkably low provision coverage ratio (defined as: stock of provisions on defaulted assets / stock of defaulted assets expressed in EAD, the Exposure at Default) for the commercial real estate portfolio that stood out in the 2011 (first) EBA Stress Test data: 4.6% as of 31 December 2010 (compare with the Danske’s 44.1%, SEB’s 37.9% or even with the Handelsbanken’s 15.8%).

To summarise, the best word for characterising the Nordea’s ROE target of 15% in a normalised macroeconomic environment is perhaps “opportunistic”. I don’t think that this target can possibly be achieved as guided by the bank so far. I’d rather expect something unexpected before such “normalisation”, some sort of more significant (structural) change. The later is of course a pure speculation.

No comments:

Post a Comment